mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-30 19:09:28 +00:00
39975613b1
Second, deeper pass over tools/gateway/hermes_cli plus first pass over the trees wave 1 missed (acp, acp_adapter, skills, computer_use, docker, dashboard, conformance, monitoring, secret_sources, hermes_state, providers). Same rubric as wave 1 (AGENTS.md test policy); security, alternation/caching invariants, issue-number regressions, and E2E kept. Real test-quality fixes found and rooted out along the way: - tests/tools/test_command_guards.py made real auxiliary-LLM HTTPS calls (DEFAULT_CONFIG smart-approval leaked in) — pinned approval mode=manual via autouse fixture: 17.4s → 0.4s. - test_model_switch_custom_providers.py / test_user_providers_model_switch.py silently probed live provider catalogs (~2s/test) — stubbed cached_provider_model_ids/provider_model_ids/fetch_api_models. - test_telegram_noise_filter.py: 15-platform copy-paste matrix over shared gateway.run logic → 3 representative platforms (55s → 3.9s). - test_gateway_shutdown.py: stop()'s 5s interrupt-deadline loop spun on MagicMock agents — interrupt.side_effect now clears _running_agents (22s → 1.0s). - test_gateway_inactivity_timeout.py poll-harness timings shrunk 3-5x (24s → 1.1s); test_mcp_stability.py backoff/SIGTERM-grace sleeps patched (15.4s → 2.5s); test_async_delegation.py negative-drain wait 5s → 0.5s. - test_telegram_init_deadline.py: loop-block margin restored to 1.0s with rationale comment — the watchdog-dump assertion needs the loop blocked well past deadline+grace under parallel load (flaked once in the 40-worker verification run at a 0.2s margin). Verification: full hermetic suite via scripts/run_tests.sh — 2,438 files, 21,718 tests passed, 0 failed, 293.9s wall. Suite totals vs original baseline: 46,820 → 19,757 test functions (−57.8%), wall 583.5s → 293.9s (−50%), subprocess CPU 13,564s → 11,623s.
88 lines
3.1 KiB
Python
88 lines
3.1 KiB
Python
"""Tests for the `log` tool_progress mode (salvage of #3459 / #3458).
|
|
|
|
`display.tool_progress: log` keeps the chat silent and appends tool-call
|
|
lines to ~/.hermes/logs/tool_calls.log via write_tool_log's rotating handler.
|
|
These tests exercise the mode's building blocks without spinning up a full
|
|
gateway run: the callback log-branch semantics and the writer coroutine.
|
|
"""
|
|
|
|
import asyncio
|
|
import queue
|
|
from datetime import datetime
|
|
|
|
import pytest
|
|
|
|
|
|
def _log_branch(log_queue, progress_queue, event_type, tool_name, preview=None):
|
|
"""Replica of the log-mode branch in gateway/run.py progress_callback."""
|
|
if log_queue is not None:
|
|
if event_type == "tool.started" and tool_name and tool_name != "_thinking":
|
|
ts = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
|
|
preview_str = f' "{preview}"' if preview else ""
|
|
log_queue.put(f"{ts} {tool_name}:{preview_str}".rstrip())
|
|
if not progress_queue:
|
|
return "returned"
|
|
return "fell-through"
|
|
|
|
|
|
class TestLogBranchSemantics:
|
|
def test_tool_started_enqueued(self):
|
|
q = queue.Queue()
|
|
assert _log_branch(q, None, "tool.started", "terminal", "ls -la") == "returned"
|
|
line = q.get_nowait()
|
|
assert "terminal" in line and "ls -la" in line
|
|
|
|
|
|
def test_thinking_not_enqueued(self):
|
|
q = queue.Queue()
|
|
_log_branch(q, None, "tool.started", "_thinking", "pondering")
|
|
assert q.empty()
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_write_tool_log_writes_and_rotates_handler(tmp_path, monkeypatch):
|
|
"""The writer coroutine drains the queue into logs/tool_calls.log."""
|
|
import gateway.run as gateway_run
|
|
|
|
monkeypatch.setattr(gateway_run, "_hermes_home", tmp_path)
|
|
|
|
log_queue: queue.Queue = queue.Queue()
|
|
log_queue.put("2026-07-02 10:00:00 terminal: \"echo hi\"")
|
|
log_queue.put("2026-07-02 10:00:01 read_file: \"foo.py\"")
|
|
|
|
# Minimal inline copy of write_tool_log wiring (the real coroutine is a
|
|
# closure inside _run_agent); exercise the same handler configuration.
|
|
import logging
|
|
from logging.handlers import RotatingFileHandler
|
|
|
|
from agent.redact import RedactingFormatter
|
|
|
|
log_dir = tmp_path / "logs"
|
|
log_dir.mkdir(parents=True, exist_ok=True)
|
|
handler = RotatingFileHandler(
|
|
log_dir / "tool_calls.log", maxBytes=5 * 1024 * 1024, backupCount=3,
|
|
encoding="utf-8",
|
|
)
|
|
handler.setFormatter(RedactingFormatter("%(message)s"))
|
|
tool_logger = logging.getLogger(f"hermes.tool_calls.test.{id(log_queue)}")
|
|
tool_logger.setLevel(logging.INFO)
|
|
tool_logger.propagate = False
|
|
tool_logger.addHandler(handler)
|
|
try:
|
|
while True:
|
|
try:
|
|
tool_logger.info("%s", log_queue.get_nowait())
|
|
except queue.Empty:
|
|
break
|
|
finally:
|
|
tool_logger.removeHandler(handler)
|
|
handler.flush()
|
|
handler.close()
|
|
|
|
content = (log_dir / "tool_calls.log").read_text(encoding="utf-8")
|
|
assert "terminal" in content
|
|
assert "read_file" in content
|
|
assert content.count("\n") == 2
|
|
await asyncio.sleep(0) # keep the asyncio marker honest
|
|
|
|
|